Objectives: ChatGPT-4 Vision (GPT-4V) is a state-of-the-art multimodal large language model (LLM) that may be queried using images. We aimed to evaluate the tool’s diagnostic performance when autonomously assessing clinical imaging studies. Materials and methods: A total of 206 imaging studies (i.e., radiography (n = 60), CT (n = 60), MRI (n = 60), and angiography (n = 26)) with unequivocal findings and established reference diagnoses from the radiologic practice of a large university hospital were accessed. Readings were performed uncontextualized, with only the image provided, and contextualized, with additional clinical and demographic information. Responses were assessed along multiple diagnostic dimensions and analyzed using appropriate statistical tests. Results: With its pronounced propensity to favor context over image information, the tool’s diagnostic accuracy improved from 8.3% (uncontextualized) to 29.1% (contextualized, first diagnosis correct) and 63.6% (contextualized, correct diagnosis among differential diagnoses) (p ≤ 0.001, Cochran’s Q test). Diagnostic accuracy declined by up to 30% when 20 images were re-read after 30 and 90 days and seemed unrelated to the tool’s self-reported confidence (Spearman’s ρ = 0.117 (p = 0.776)). While the described imaging findings matched the suggested diagnoses in 92.7%, indicating valid diagnostic reasoning, the tool fabricated 258 imaging findings in 412 responses and misidentified imaging modalities or anatomic regions in 65 images. Conclusion: GPT-4V, in its current form, cannot reliably interpret radiologic images. Its tendency to disregard the image, fabricate findings, and misidentify details, especially without clinical context, may misguide healthcare providers and put patients at risk. Key Points: Question Can Generative Pre-trained Transformer 4 Vision (GPT-4V) interpret radiologic images—with and without clinical context? Findings GPT-4V performed poorly, demonstrating diagnostic accuracy rates of 8% (uncontextualized), 29% (contextualized, most likely diagnosis correct), and 64% (contextualized, correct diagnosis among differential diagnoses). Clinical relevance The utility of commercial multimodal large language models, such as GPT-4V, in radiologic practice is limited. Without clinical context, diagnostic errors and fabricated findings may compromise patient safety and misguide clinical decision-making. These models must be further refined to be beneficial.

Visual large language models (VLLMs) are discussed as potential tools for assisting radiologists in image interpretation, yet their clinical value remains unclear. This study provides a systematic and comprehensive comparison of general-purpose and biomedical VLLMs in radiology. We evaluated 180 representative clinical images with validated reference diagnoses (radiography, CT, MRI; 60 each) using seven VLLMs (ChatGPT-4o, Gemini 2.0, Claude Sonnet 3.7, Perplexity AI, Google Vision AI, LLaVA-1.6, LLaVA-Med-v1.5). Each model interpreted the image without and with clinical context. Mixed-effects logistic regression models assessed the influence of model, modality, and context on diagnostic performance and hallucinations (fabricated findings or misidentifications). Diagnostic accuracy varied significantly across all dimensions (p ≤ 0.001), ranging from 8.1% to 29.2% across models, with Gemini 2.0 performing best and LLaVA performing weakest. CT achieved the best overall accuracy (20.7%), followed by radiography (17.3%) and MRI (13.9%). Clinical context improved accuracy from 10.6% to 24.0% (p < 0.001) but shifted the model to rely more on textual information. Hallucinations were frequent (74.4% overall) and model-dependent (51.7–82.8% across models; p ≤ 0.004). Current VLLMs remain diagnostically unreliable, heavily context-biased, and prone to generating false findings, which limits their clinical suitability. Domain-specific training and rigorous validation are required before clinical integration can be considered.

OpenAI’s GPT-4 with Advanced Data Analysis autonomously performed basic statistical analyses and built machine learning–based models to evaluate a large chest radiography dataset, performing similarly to its manually crafted counterpart.

While providing the reference imaging modality for joint pathologies, MRI is focused on morphology and static configurations, thereby not fully exploiting the modality’s diagnostic capabilities. This study aimed to assess the diagnostic value of stress MRI combining imaging and loading in differentiating partial versus complete anterior cruciate ligament (ACL)-injury. Ten human cadaveric knee joint specimens were subjected to serial imaging using a 3.0T MRI scanner and a custom-made pressure-controlled loading device. Emulating the anterior-drawer test, joints were imaged before and after arthroscopic partial and complete ACL transection in the unloaded and loaded configurations using morphologic sequences. Following manual segmentations and registration of anatomic landmarks, two 3D vectors were computed between anatomic landmarks and registered coordinates. Loading-induced changes were quantified as vector lengths, angles, and projections on the x-, y-, and z-axis, related to the intact unloaded configuration, and referenced to manual measurements. Vector lengths and projections significantly increased with loading and increasing ACL injury and indicated multidimensional changes. Manual measurements confirmed gradually increasing anterior tibial translation. Beyond imaging of ligament structure and functionality, stress MRI techniques can quantify joint stability to differentiate partial and complete ACL injury and, possibly, compare surgical procedures and monitor treatment outcomes.

Background: Traumatic cartilage injuries predispose articulating joints to focal cartilage defects and, eventually, posttraumatic osteoarthritis. Current clinical-standard imaging modalities such as morphologic MRI fail to reliably detect cartilage trauma and to monitor associated posttraumatic degenerative changes with oftentimes severe prognostic implications. Quantitative MRI techniques such as T2 mapping are promising in detecting and monitoring such changes yet lack sufficient validation in controlled basic research contexts. Material and Methods: 35 macroscopically intact cartilage samples obtained from total joint replacements were exposed to standardized injurious impaction with low (0.49 J, n = 14) or high (0.98 J, n = 14) energy levels and imaged before and immediately, 24 h, and 72 h after impaction by T2 mapping. Contrast, homogeneity, energy, and variance were quantified as features of texture on each T2 map. Unimpacted controls (n = 7) and histologic assessment served as reference. Results: As a function of impaction energy and time, absolute T2 values, contrast, and variance were significantly increased, while homogeneity and energy were significantly decreased. Conclusion: T2 mapping and texture feature analysis are sensitive diagnostic means to detect and monitor traumatic impaction injuries of cartilage and associated posttraumatic degenerative changes and may be used to assess cartilage after trauma to identify “cartilage at risk”.

Quantitative MRI techniques such as T2 and T1ρ mapping are beneficial in evaluating knee joint pathologies; however, long acquisition times limit their clinical adoption. MIXTURE (Multi-Interleaved X-prepared Turbo Spin-Echo with IntUitive RElaxometry) provides a versatile turbo spin-echo (TSE) platform for simultaneous morphologic and quantitative joint imaging. Two MIXTURE sequences were designed along clinical requirements: “MIX1”, combining proton density (PD)-weighted fat-saturated (FS) images and T2 mapping (acquisition time: 4:59 min), and “MIX2”, combining T1-weighted images and T1ρ mapping (6:38 min). MIXTURE sequences and their reference 2D and 3D TSE counterparts were acquired from ten human cadaveric knee joints at 3.0 T. Contrast, contrast-to-noise ratios, and coefficients of variation were comparatively evaluated using parametric tests. Clinical radiologists (n = 3) assessed diagnostic quality as a function of sequence and anatomic structure using five-point Likert scales and ordinal regression, with a significance level of α = 0.01. MIX1 and MIX2 had at least equal diagnostic quality compared to reference sequences of the same image weighting. Contrast, contrast-to-noise ratios, and coefficients of variation were largely similar for the PD-weighted FS and T1-weighted images. In clinically feasible scan times, MIXTURE sequences yield morphologic, TSE-based images of diagnostic quality and quantitative parameter maps with additional insights on soft tissue composition and ultrastructure.